INDEX

Explanations

instances of the word "that" and expressions related to clarification or specification

oai_token-act-pair · gpt-4o-mini Triggered by @bot

that clauses

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ſte

-0.56

 ſtre

-0.51

 ſch

-0.50

 ſche

-0.50

LEM

-0.49

vro

-0.49

 Verdun

-0.49

 myſelf

-0.47

 itſelf

-0.47

ACHUSET

-0.47

POSITIVE LOGITS

Билгалдахарш

0.53

jsxFileName

0.49

ortunadamente

0.47

 nahilalakip

0.47

brigens

0.46

 الحره

0.45

ValueStyle

0.45



0.44

__((

0.44

脚注の使い方

0.43

Activations Density 0.054%

instances of the word "that" and expressions related to clarification or specification

that clauses

No Comments

No Known Activations